Fall Detection Algorithm using the Fall Feature Parameters Extracted from Video and Accelerometer Sensor

A newly developed fall detection algorithm applied to a simple threshold method and hidden Markov model (HMM) is introduced. The algorithm is based on the fall feature parameters extracted from the video of surrounding environment using overhead camera and accelerometer data of the subject's bo...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Information Institute (Tokyo). Information Ročník 19; číslo 2; s. 615
Hlavní autoři: Yu, YunSeop, Kim, Sang-Hoon
Médium: Journal Article
Jazyk:angličtina
Vydáno: Koganei International Information Institute 01.02.2016
Témata:
Accelerometers
Activities of daily living
Algorithms
Cameras
Feature extraction
Learning
Markov models
Mathematical models
Parameters
Probability distribution
Sensors
Thresholds
Wearable computers
ISSN:1343-4500, 1344-8994
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:A newly developed fall detection algorithm applied to a simple threshold method and hidden Markov model (HMM) is introduced. The algorithm is based on the fall feature parameters extracted from the video of surrounding environment using overhead camera and accelerometer data of the subject's body. The previously proposed fall feature parameters of video and accelerometer are used to distinguish between a fall and normal activities of daily living, and they are applied to the previously developed fail detection algorithm combining the simple threshold method and HMM. Compared with the authors' previously reported results of separated wearable device and camera, the results obtained with the proposed algorithm are better with a sensitivity of 99.69%, a specificity of 99.69%, and an accuracy of 99.69%.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:1343-4500
1344-8994